diff options
author | Claudio Freire <klaussfreire@gmail.com> | 2015-07-20 22:53:24 -0300 |
---|---|---|
committer | Michael Niedermayer <michael@niedermayer.cc> | 2015-07-27 19:13:48 +0200 |
commit | 59216e0525a58714be4207be6ae8744750e62867 (patch) | |
tree | 8c5b41b3471b08e17396c159001da6deff7ce347 /libavcodec/psymodel.h | |
parent | c8c86b8f9b8c166633a7324c8646f38866801b88 (diff) | |
download | ffmpeg-59216e0525a58714be4207be6ae8744750e62867.tar.gz |
AAC Encoder: clipping avoidance
Avoid clipping due to quantization noise to produce audible
artifacts, by detecting near-clipping signals and both attenuating
them a little and encoding escape-encoded bands (usually the
loudest) rounding towards zero instead of nearest, which tends to
decrease overall energy and thus clipping.
Currently fate tests measure numerical error so this change makes
tests using asynth (which are near clipping) report higher error
not less, because of window attenuation. Yet, they sound better,
not worse (albeit subtle, other samples aren't subtle at all).
Only measuring psychoacoustically weighted error would make for
a representative test, so that will be left for a future patch.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Diffstat (limited to 'libavcodec/psymodel.h')
-rw-r--r-- | libavcodec/psymodel.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/libavcodec/psymodel.h b/libavcodec/psymodel.h index 2e3ab911e4..e9be1f6fa5 100644 --- a/libavcodec/psymodel.h +++ b/libavcodec/psymodel.h @@ -66,6 +66,7 @@ typedef struct FFPsyWindowInfo { int window_shape; ///< window shape (sine/KBD/whatever) int num_windows; ///< number of windows in a frame int grouping[8]; ///< window grouping (for e.g. AAC) + float clipping[8]; ///< maximum absolute normalized intensity in the given window for clip avoidance int *window_sizes; ///< sequence of window sizes inside one frame (for eg. WMA) } FFPsyWindowInfo; |